Okay, finally took the time to package up the app and upload it to Google Play.

Here's the app's page on the Play Store (with screenshots): Lucid Link - Google Play Store
And here's the source code for it: https://github.com/Venryx/LucidLink

The app and this technique are still early in development, so this thread is a "beta" version of it, for people who want to try it themselves before I've worked out each of the components thoroughly.

Before we get onto the technique, here's a bit of background on it:
Spoiler for Technique background and success rates:

I'll make a nicer and more comprehensive tutorial later on, but for now, this is the summary for the technique, as I'm currently doing it, which I dubbed last night the RVP (random voice prompt) technique. It's divided into two parts: the technical setup, and the nightly practice.

Technical setup

1) Launch the app, and open the "Tools>RVP Technique" tab.

2) Temporarily change the voice-prompt interval to 10 seconds (for volume adjustment), then toggle the "enabled" switch on.

3) Adjust the volume so that it's loud enough to gather attention (reminding you to make your mental response), but not enough to prevent sleep. The spoiler below describes how to replicate the volume I've been using, but you can skip it if you'd rather adjust it all yourself anyway.
Spoiler for Volume adjustment:

4) Change the voice-prompt interval so that it's as frequent as it can be without you "burning out" -- where you ignore the prompts or are substantially annoyed by them. In other words, make sure the prompts are still rare and "interesting enough" that you can comfortably "listen in" and process them the whole time you're (consciously) falling asleep. This also requires much experimentation, but here's what I've been using so far (the last month or so): every 120 seconds

5) Customize the range of tones/pitches to use for the text-to-speech voice. Currently I have it cover all the way from 1% to 150% (150%+ becomes hard to recognize), but I haven't tried other ranges much yet, so I don't know how much it affects the technique's effectiveness. I do think some variation should occur though, as otherwise you'd probably become too accustomed to the sound of the voice. (I do know that the short time before I added pitch variation, the technique didn't seem to work much.) Anyway, feel free to experiment, though I suggest allowing for at least an 80% or so range for pitch variation. (and btw, at first the low tones can sound kind of scary/weird, but you get used to it after several days)

6) Augment the list of voice-prompt names with the names of people you know personally and who are significant to you. This will include most of your friends, your family, some of your relatives probably, other members on DreamViews, and any dream guides or the like from your dream worlds. These names will be spoken by the voice prompts, to make them more interesting and connect you with them more meaningfully. Also! Make sure you include your own name in that list. It has an interesting mental effect when you hear a robot voice in the night randomly saying your name followed by a set of random digits.

Great! Now you're set up and can focus on what you're supposed to do each night while falling asleep.

Nightly practice

The technique has worked some for my long/night sleep, though as seen by the stats it's much less reliable than if it's combined with WBTB. Thus, I definitely recommend doing WBTB if you're intent on the technique succeeding quickly (and with full-fledged lucids instead of "weak" ones, as are much more common for long/night sleeps). But either way, you end up doing the same procedure whenever you enter bed, so you can just do it with WBTB whenever it's convenient.

1) Start the app, and enable the RVP tool. (making sure the volume is set to the value determined earlier, and the device/speaker is in position)

2) Begin "waiting". What to do while you're waiting has been the main thing I've experimented with over the last month. Interestingly, it seems a good variety of things have worked, including:
Spoiler for Activities while waiting:

3) Whenever a voice prompt occurs, listen carefully for the number that is spoken. The number is randomly generated, but in such a way that either all its digits will be even, or all its digits will be odd. Now this is important because it lets you use it as a "passive reality-check" to evaluate whether you're in a dream/false-awakening without having to actually move or anything. So whenever a number is spoken, check the digits to make sure that they match the "all even or all odd" criteria. If not, then congrats! You've caught your subconscious trying (but failing) to imitate the actual app, and you must be in a lucid dream. If so, then mentally speak to yourself the state of those numbers; ie, if they were all even, mentally speak/think the word "even", and if they were all odd, mentally speak/think the word "odd".

4) (optional) You can also try something I've been trying recently, which is reciting a "quick list" of your top four dream goals just after your voice-prompt response above. I suspect this reinforces your intention to lucid dream, as well as making it easier to remember your goals once you actually enter one. If you want, you can also combine this with a brief visualization of those four dream goals, as visual memories can be easier to access relationally while in-dream. (I envision my dream goals as being physically stacked on each other in interesting ways. For example, if my goals are "talk to my coworker", "get a massage", "ask a pencil to draw a picture", and "call down a storm of meteors", I would recite my quick-list as "coworker, massage, pencil, meteor", and/or visualize it as my coworker, with a couch on her head, a pencil on the couch, and a meteor balanced on the pencil.)

That's the bulk of it. There's more to add in way of things I've learned along the way, and some pitfalls (such as desensitizing to the prompts and no longer giving them significant attention, which reduced the effectiveness for a day or two), but I'll speak on those things later when I've experimented more myself and am ready to package it up again for "version 1". (this is still a beta)

Let me know if you have any questions, and/or if you want to participate in a public evaluation of the technique. I'll be doing so myself here each night -- posting my results, describing what I learn, and making changes to the technique (and app) along the way.